ADF Pipeline Error:
ADF Pipeline Error 2302

What does this error mean?

Hadoop job submitted to HDI cluster failed with non-zero exit code; check stderr log path or Ambari UI for the job ID.

Common causes

• 1HDInsight cluster service in alert/unhealthy state (check Ambari dashboard)
• 2Insufficient YARN memory or cores for the Spark/MapReduce job, causing delays or failure
• 3Job not submitted to YARN (missing --master yarn) so it never got proper resource allocation
• 4Application-level error in the job code itself (logic, data, or dependency failure)

How to fix it

1. 1Open the Ambari UI on the HDI cluster and check the alert section on the dashboard — make sure all services are still running.
2. 2Review YARN memory and cores; if utilization is high, scale up the cluster so it can accommodate the Spark/Hadoop job.
3. 3Verify the job was submitted to YARN (use --master yarn if not) and run a sample MapReduce/Spark job on the HDI backend to confirm cluster health.
4. 4Collect application logs with `yarn logs -applicationId <Your_Application_ID>` and gather YARN ResourceManager logs from /var/log/hadoop-yarn/yarn, plus start/launch timestamps.
5. 5If unresolved, contact the Azure HDInsight support team with the collected logs, application ID and timestamps.

Frequently asked questions

Other configuration errors